He Yigang

Steganography based on wavelet transform and modulus function

Abstract In order to provide larger capacity of the hidden secret data while maintaining a good visual quality of stego-image, in accordance with the visual property that human eyes are less sensitive to strong texture, a novel steganographic method based on wavelet and modulus function is presented. First, an image is divided into blocks of prescribed size, and every block is decomposed into one-level wavelet. Then, the capacity of the hidden secret data is decided with the number of wavelet coefficients of larger magnitude. Finally, secret information is embedded by steganography based on modulus function. From the experimental results, the proposed method hides much more information and maintains a good visual quality of stego-image. Besides, the embedded data can be extracted from the stego-image without referencing the original image.

Diagonal recurrent neural network based on-line stator winding turn fault detection for induction motors

The main limitation of feed-forward neural network based modeling methods for stator winding turn fault detection is its poor dynamical processing capability. To solve this problem, a diagonal recurrent neural network based on-line turn fault detection approach for induction motors is presented in this paper. Two diagonal recurrent neural networks are employed to detect turn fault. One is used to estimate the fault severity, the other is used to determine the exact number of fault turns. In order to make the diagonal recurrent neural network model more simple and accurate, an adaptive dynamic back propagation algorithm is proposed to determine the optimum number of the hidden layer neurons. Experiments are carried out on a special rewound laboratory induction motor, the results show that the diagonal recurrent neural network based diagnosis model determines the shorted turns exactly, and is more effective than the forward neural network based diagnosis model under the condition of detecting a slowly developing turn fault.

Fuzzy Neural Network based On-line Stator Winding Turn Fault Detection for Induction Motors

A fuzzy neural network based on-line turn fault detection approach for induction motors is proposed in this paper. B-spline membership fuzzy neural network is employed to detect turn fault, since the selection of the weighting factors, the knot positions and the control points of the B-spline membership fuzzy-neural networks is crucial to obtaining good approximation for complex nonlinear systems, a genetic algorithm with an efficient search strategy is developed to optimize network parameters. Based on it, Experiments are carried out on a special rewound laboratory induction motor, the results show fuzzy neural network based diagnosis model determines the shorted turns exactly, and is more effective than the parameters estimation method under the condition of detecting a slowly developing turn fault.

On the application of artificial neural networks to fault diagnosis in analog circuits with tolerances

This paper proposes a method for analog fault diagnosis by adopting neural networks. The primary focus of the paper is to provide robust diagnosis using a mechanism to deal with the problem of component tolerances and to reduce the testing time. The proposed approach is based on the k-fault diagnosis method and artificial backward propagation neural network, which is shown to be capable of robust diagnosis of analog circuits with tolerances.

A novel method for fault diagnosis of analog circuits based on WP and GPNN

A novel method for fault diagnosis of analog circuits with tolerance based on wavelet packet (WP) decomposition and probabilistic neural networks using genetic algorithm (GPNN) is proposed in this paper. The fault feature vectors are extracted after feasible domains on the basis of WP decomposition of responses of a circuit being solved. Then by fusing various uncertain factors into probabilistic operations, GPNN methods to diagnose faults are proposed whose parameters and structure obtained form genetic optimisations resulting in best detection of faults. Finally, simulations indicated that GPNN classifiers are correct 7% more than BPNN of the test data associated with our sample circuits.

A neural-based nonlinear L 1-norm optimization algorithm for diagnosis of networks

Based on exact penalty function, a new neural network for solving the L1-norm optimization problem is proposed. In comparison with Kennedy and Chua’s network(1988), it has better properties.Based on Bandler’s fault location method(1982), a new nonlinearly constrained L1-norm problem is developed. It can be solved with less computing time through only one optimization processing. The proposed neural network can be used to solve the analog diagnosis L1 problem. The validity of the proposed neural networks and the fault location L1 method are illustrated by extensive computer simulations.

Simulation of grain boundary effect on characteristics of ZnO thin film transistor by considering the location and orientation of grain boundary

The grain boundaries (GBs) have a strong effect on the electric properties of ZnO thin film transistors (TFTs). A novel grain boundary model was developed to analyse the effect. The model was characterized with different angles between the orientation of the grain boundary and the channel direction. The potential barriers formed by the grain boundaries increase with the increase of the grain boundary angle, so the degradation of the transistor characteristics increases. When a grain boundary is close to the drain edge, the potential barrier height reduces, so the electric properties were improved.

Experimental and statistical analysis of blind spots for UHF RFID portal applications

The radiation field of a typical UHF RFID portal application was measured using our developed LabVIEW based automatic test platform. Path loss exponent and standard deviation of Log-normal shadowing model were obtained through MMSE criteria from measured data. Rician and Rayleigh empirical models were also introduced to predict the probability of blind spots. And fade margins were successfully compared for different blind spots probability.

Fuzzy Neural Network based Predictive Control for Active Power Filter

Fuzzy neural network based predictive control of active power filter is presented in this paper. In the scheme, B-spline membership fuzzy neural network is employed to predict future harmonic compensating current. In order to make the predictive model compact and accurate, a genetic algorithm with an efficient search strategy is developed to optimize the weighting factors, the knot positions and the control points of the B-spline membership fuzzy neural networks. Based on the model output, branch-and-bound optimization method is adopted to produce proper value of control vector. This control vector is adequately modulated by means of a space vector PWM modulator which generates proper gating patterns of the inverter switches to maintain tracking of reference current. Fuzzy neural network based predictive algorithm is used in internal model control scheme to compensate for process disturbances, measurement noise and modeling errors. Experiment on an actual system is implemented. The results show fuzzy neural network based predictive control eliminates supply current and is more effective than PI control.

A fault identification approach for analog circuits using fuzzy neural network mixed with genetic algorithms

A fault identification approach for nonlinear analogue systems is presented. A fuzzy neural network is developed based on the improving fuzzy weighted reasoning method. The training of network weights and optimization of membership functions are conducted employing genetic algorithms. Fuzzy rules can be realized through the refresh of the weights of the neural network. The availability of the method is examined by simulated test examples.